Lectures related to Controlled Stochastic Processes

Explores Dynamic Programming for optimal control, focusing on stability, stationary policy, and recursive solutions.

Explores diffusion models, focusing on generating samples from a distribution and the importance of denoising in the process.

Introduction to Reinforcement Learning: Key Concepts and Applications

Introduces reinforcement learning, covering its definitions, applications, and theoretical foundations, while outlining the course structure and objectives.

Markov Decision Processes: Foundations of Reinforcement Learning

Covers Markov Decision Processes, their structure, and their role in reinforcement learning.

Optimal Transport: Kantorovich Duality

Covers optimal transport and Kantorovich duality in real-life distribution problems.

Approximation Algorithms

Covers approximation algorithms for optimization problems, LP relaxation, and randomized rounding techniques.

Controlled Stochastic Processes

Explores controlled stochastic processes, dynamic programming, and the Machine Replacement Problem.

Generalization Error

Explores generalization error in machine learning, focusing on data distribution and hypothesis impact.

Set Cover: Integrality Gap

Explores the integrality gap concept in set cover and multiplicative weights algorithms.

Optimization Methods: Theory Discussion

Explores optimization methods, including unconstrained problems, linear programming, and heuristic approaches.

Sobolev Spaces in Higher Dimensions

Explores Sobolev spaces in higher dimensions, discussing derivatives, properties, and challenges with continuity.

Dynamic Programming: Optimal Decision Making

Explores dynamic programming for optimizing decision-making processes over time, using real-world examples like oil extraction and stock trading.

Value Iteration Acceleration: PID and Operator Splitting

Explores accelerating the Value Iteration algorithm using control theory and matrix splitting techniques to achieve faster convergence.

Sparsest Cut: Bourgain's Theorem

Explores Bourgain's theorem on sparsest cut in graphs, emphasizing semimetrics and cut optimization.

Infinite-Horizon LQ Control: Solution & Example

Explores Infinite-Horizon Linear Quadratic (LQ) optimal control, emphasizing solution methods and practical examples.

Deep Learning Modus Operandi

Explores the benefits of deeper networks in deep learning and the importance of over-parameterization and generalization.

Finite Element Method: Weak Solutions

Covers weak solutions in the finite element method, emphasizing continuity and the Cauchy-Schwarz inequality.

Nonlinear Model Predictive Control

Explores Nonlinear Model Predictive Control, covering stability, optimality, pitfalls, and examples.

Markov Chains: Transition Probabilities

Explores Markov chains, transition matrices, distribution, and random walks.

Asset Selling: Optimal Revenue Policy

Explores asset selling dynamics, optimal revenue policy, acceptance thresholds, and commodity price impact.